xrogzu / json-rules-engine

JSON Rule Engine

What does it do?

Pass a Rule and Data to engine and provide synchronous response on whether the Data satisfies the Rules.

Installation

    npm install @aipeel/json-rules-engine

Usage

Step 1: Define Rules JSON

1. Rules JSON must be an object

2. Must start with AND or OR which takes array of rule

   1. Single Rule Criteria

   {
       AND: [ Criteria1 ]
   }
   

   2. Multiple Rule Criteria --> Criteria 1 and 2 must be satisfied

   {
       AND: [ Criteria1, Criteria2 .. ]
   }
   

   3. Multiple Levels of Rules --> Criteria 1 & 2 must be satisfied and either Criteria3 or Criteria4 must be satisfied

   {
       AND: [
               {
                   AND: [ Criteria1, Criteria2 ...]
               },
               {
                   OR: [Criteria3, Criteria4 ..]
               }
           ]
   }
   

3. Each Criteria must be an object

   1. The key in criteria is an operator and value defines the metadata required for the operator to work
      
      Examples of Criteria

      { 
          "between": { 
                          "min": "10", 
                          "max": "20", 
                          "attribute": "discount" 
                      }
      }

      { 
          "in": { 
                  "value": ["C","Javascript","Python"], 
                  "attribute": "lang"
              }
      }

   2. List of all criteria given below

      Operator	Operator Property	Description	Example
      between	min, max, attribute	Check if the property mentioned in attribute has value between min and max	{ "between" : { "min" : "10" , "max" : "20" , "attribute" : "discount" }}
      lessThan	value, attribute	Check if the property mentioned in attribute has number less than value	{ "lessThan" : { "value" : "80" , "attribute" : "price" } }
      lessThanEqual	value, attribute	Check if the property mentioned in attribute has number less than or equal value	{ "lessThanEqual" : { "value" : "80" , "attribute" : "price" } }
      greaterThan	value, attribute	Check if the property mentioned in attribute has number greater than value	{ "greaterThan" : { "value" : "80" , "attribute" : "price" } }
      greaterThanEqual	value, attribute	Check if the property mentioned in attribute has number greater than or equal value	{ "greaterThanEqual" : { "value" : "80" , "attribute" : "price" } }
      equal	value, attribute	Check if the property mentioned in attribute has value equal to value provided	{ "equal" : { "value" : "80" , "attribute" : "price" } }
      contains	value, attribute	Check if the property mentioned in attribute contains the value given.	{ "contains" : { "value" : "great" , "attribute" : "desc" }}
      in	value, attribute	Check if the property mentioned in attribute is available in value array provided.	{ "in" : { "value" : [ "C" , "Javascript" , "Python" ], "attribute" : "lang" }}

   3. A Rules JSON could be defined as

   { 
       "AND": [
           { "lessThan": {"value": "80", "attribute": "price"} },
           { "greaterThanEqual": {"value": "20", "attribute": "price"} },
           { "between": { "min": "10", "max": "20", "attribute": "discount" }}
       ]
   }

   {
       "OR":[
           { "contains": { "value": "great", "attribute": "desc" }},
           { "contains": { "value": "works", "attribute": "features"}},
           { "in": { "value": ["C","Javascript","Python"], "attribute": "lang"}},
       ]
   }

4. Rules can be recursively nested as shown below

   {
       "AND": [
               { 
                   "AND": [
                       { "lessThan": {"value": "80", "attribute": "price"} },
                       { "greaterThanEqual": {"value": "20", "attribute": "price"} },
                       { "between": { "min": "10", "max": "20", "attribute": "discount" }}
                   ]
               },
               {
                   "OR":[
                       { "contains": { "value": "great", "attribute": "desc" }},
                       { "contains": { "value": "works", "attribute": "features"}},
                       { "in": { "value": ["C","Javascript","Python"], "attribute": "lang"}},
                   ]
               }
           ]
   }

Step2: Pass Data JSON and Rule JSON to Engine

    const Engine = require("@aipeel/json-rules-engine")

    Engine.apply(data,rule) // returns boolean - true or false

Complete end-to-end example below for reference

const Engine = require("@aipeel/json-rules-engine")

const rule = {
                    "AND": [
                            { 
                                "AND": [
                                    { "lessThan": {"value": "80", "attribute": "price"} },
                                    { "greaterThanEqual": {"value": "20", "attribute": "price"} },
                                    { "between": { "min": "10", "max": "20", "attribute": "discount" }}
                                ]
                            },
                            {
                                "OR":[
                                    { "contains": { "value": "great", "attribute": "desc" }},
                                    { "contains": { "value": "works", "attribute": "features"}},
                                    { "in": { "value": ["C","Javascript","Python"], "attribute": "lang"}},
                                ]
                            }
                        ]
                    }

let data = {
    id: 'ID1001',
    price: 50,
    discount: 10,
    desc: "This is a great product for developers",
    features: ["No Dependencies", "Open to Extension", "Synchronous", "works"],
    tags: ["json", "rule", "engine"],
    lang: "Javascript"
}

// example 1
console.log(`Does the data conform to rule? --> ${Engine.apply(data,rule)?"Yes":"No"}`);

//example 2
data.price = 100;
console.log(`Does the data conform to rule? --> ${Engine.apply(data,rule)?"Yes":"No"}`);

//example 3
data.price = 50;
data.discount = 21;
console.log(`Does the data conform to rule? --> ${Engine.apply(data,rule)?"Yes":"No"}`);

Output

Does the data conform to rule? --> Yes
Does the data conform to rule? --> No
Does the data conform to rule? --> No